##Table of Contents
-
Nipype installation: Follow nipype installation guide or this one.
-
Download MVPA Analysis Open the terminal and run
git clone https://github.com/labmukamel/mvpa_analysis.git
Notes:
-
You might run into problems with running FSL in pycharm. Add the following to pycharm.sh (In: [pycharm installation]/bin/pycharm.sh):
./etc/fsl/5.0/fsl.sh
The task order file can be general for all subjects and for a specific subject. In order to use it for specific subjects we can add a file task_order_SubjectName.txt in the save directory as the task_order.txt
task001_run001
task001_run002
task001_run003
task002_run001
MVPA_run1 task001_run001
MVPA_run2 task001_run001
MVPA_run3 task001_run001
MVPA_run4 task001_run001
MVPA_run5 task002_run001
{"AzOr": 1}
Onset Rsponse StimVar Stimulus ResponseTime
0 1 2 incongruent 1095
10 1 2 incongruent 988
20 1 1 congruent 591
30 1 1 congruent 499
40 1 2 incongruent 719
52 1 1 congruent 544
64 1 1 congruent 436
76 1 2 incongruent 470
task001 cond001 congruent_correct
task001 cond002 congruent_incorrect
task001 cond003 incongruent_correct
task001 cond004 incongruent_incorrect
task001 incongruent_vs_congruent -1 -1 1 1
task001 incorrect_vs_correct -1 1 -1 1
task001 incongruent_vs_congruent_correct -1 0 1 0
task001 incorrect_vs_correct_incongruent 0 0 -1 1
def SubjectDir(subject_code, path, raw_path, task_order, task_mapping)"Parameters
- subject_code: The subject code
- path: Output path in openfmri format
- raw_path: Contains the dicom folders
- task_order.txt: List from to task_order.txt
- task_mapping: List from task_mapping.txt
When we initiate the class, it prepares the subject subdirectories as followed:
- Creates BOLD, anatomy, model, masks folders
- Creates BOLD/taskxxx_runxxx
- Creates modelxxx/onsets/taskxxx
- Creates Mask/anatomytaskxxx_runxxx
- Creates behav/taskxxx_runxxx
And converts the DICOM files to NIFTY
If the path exists(we already created the openfmri structure) then we read it and don't overwrite
def OpenFMRIData(data_dir, raw_data_dir, study_name):Parameters
- data_dir: Target openfmri directory
- raw_data_dir: Root raw directory (Only the path to the raw directory (in raw/study_name/subxxx))
- study_name: The name of the study folder containing relevant subjects
def create_subject_dir(subject_name, overwrite = False, create_behav_dict = None):Creates the openfmri structure by creating SubjectDir
Parameters
- subject_name = (string)
- overwrite = States whether we overwrite existing directory
- create_behav_dict = dictionary[2] {'func': function that creates the conditions of the models, 'behav': behavioural path}
Returns:
- SubjectDir Object
def load_subject_dir(**kwargs):Creates or uses the openfmri structure by creating SubjectDir
- Creates the new directories only when given subname!
- Uses current structure when given subcode/subname and the subject_mapping file exists in the study dir
Parameters
- subcode = (number) or subname = (string)
- create_behav_dict = dictionary[2] {'func': function that creates the conditions of the models, 'behav': behavioural path}
- The function will receive: subjectdir, onset_dirs,behavioural_path
Returns:
- SubjectDir Object
def OpenFMRIAnalyzer(fmri_data, subjects):Parameters
- fmri_data: OpenFMRIData object
- subjects: Can either be a list of names or a list of SubjectDir
def extract_brain(self, subject, overwrite = False, f=0.5, g=-0.1):Outputs:
- mask/anatomy/brain.nii.gz = masked image from fsl view
- highres001_brain.nii.gz = The extracted brain image
Parameters
- subject = Subject Dir object
- overwrite = States whether or not to overwrite the image
- f = fractional intensity threshold
- g = vertical gradient in fractional intensity threshold (-1, 1)
Returns
- Path to anatomy/highres001_brain.nii.gz
def estimate_bias_field(self, subject, brain_image, overwrite=False)- Before running FAST an image of a head should first be brain-extracted, using BET. The resulting brain-only image can then be fed into FAST.
Outputs:
- highres001_bias.nii.gz = output estimated bias field
- highres001_restore.nii.gz = output restored image (bias-corrected image)
Parameters
- subject = Subject Dir object
- brain_image = Path to the image after it was extracted by BET
- overwrite = False
Returns:
- Path to 'highres001.nii.gz' file
def slice_time_correction(self,subject,time_repetition = 2):Outputs:
- BOLD/run_name/bold_stc.nii.gz
Parameters
- subject = Subject Dir object
- time_repetition = TR of data
def functional_smoothing(self,subject,fwhm,brightness_threshold,use_median = True):
def anatomical_smoothing(self,subject,fwhm,brightness_threshold,use_median = True):Using SUSAN to perform smoothing
Outputs:
- BOLD/run/bold_smooth.nii.gz
- anatomical/highres001_smooth.nii.gz
Parameters
- subject = Subject Dir object
- fwhm = fwhm of smoothing, in mm, gets converted using sqrt(8*log(2))
- brightness_threshold = brightness threshold and should be greater than noise level and less than contrast of edges to be preserved.
- use_median = whether to use a local median filter in the cases where single-point noise is detected
def motion_correction(subject, merge_task_runs=False)Uses MCFLIRT to run motion correction on the bold image
Outputs:
- bold_mcf.nii.gz = The image after motion correction for each functional folder
Parameters
- subject = Subject Dir object
- merge_task_runs = if true - Merges the files before motion correction and after it's done we split them back
def anatomical_registration(subject,brain_image,standard_image_name='MNI152_T1_2mm_brain.nii.gz'):- Runs FLIRT on the brain extracted anatomy image with MNI152_T1_2mm_brain.nii.gz as reference
- Runs FNIRT on the anatomy image
Outputs:
-
anatomy/reg/highres2standard.nii.gz
-
anatomy/reg/highres2standard.mat
-
anatomy/reg/highres2highres_jac
-
anatomy/reg/highres2standard_warp.nii.gz
Parameters
- subject = Subject Dir object
- standard_image_name = The anatomy reference image
def functional_segmentation(subject):Outputs:
- masks/run_name/grey.nii.gz
Parameters
- subject = Subject Dir object
def segmentation(subject):Outputs:
- masks/anatomy/grey.nii.gz
Parameters
- subject = Subject Dir object